Slashdot Mirror
D-Wave's 2,000-Qubit Quantum Annealing Computer Now 1,000x Faster Than Previous Generation (tomshardware.com)
An anonymous reader quotes a report from Tom's Hardware: D-Wave, a Canadian company developing the first commercial "quantum computer," announced its next-generation quantum annealing computer with 2,000 qubits, which is twice as many as its previous generation had. One highly exciting aspect of quantum computers of all types is that beyond the seemingly Moore's Law-like increase in number of qubits every two years, their performance increases much more than just 2x, unlike with regular microprocessors. This is because qubits can hold a value of 0, 1, or a superposition of the two, making quantum systems able to deal with much more complex information. If D-Wave's 2,000-qubit computer is now 1,000 faster than the previous 1,000-qubit generation (D-Wave 2X), that would mean that, for the things Google tested last year, it should now be 100 billion times faster than a single-core CPU. The new generation also comes with control features, which allows users to modify how D-Wave's quantum system works to better optimize their solutions. These control features include the following capabilities: The ability to tune the rate of annealing of individual qubits to enhance application performance; The ability to sample the state of the quantum computer during the quantum annealing process to power hybrid quantum-classical machine learning algorithms that were not previously possible; The ability to combine quantum processing with classical processing to improve the quality of both optimization and sampling results returned from the system. D-Wave's CEO, Vern Brownell, also said that D-Wave's quantum computers could also be used for machine learning task in ways that wouldn't be possible on classical computers. The company is also training the first generation of programmers to develop applications for D-Wave quantum systems. Last year, Google said that D-Wave's 1,000 qubit computer proved to be 100 million times faster than a classical computer with a single core: "We found that for problem instances involving nearly 1,000 binary variables, quantum annealing significantly outperforms its classical counterpart, simulated annealing. It is more than 10^8 times faster than simulated annealing running on a single core," said Hartmut Neven, Google's Director of Engineering.
at doing nothing
What can you actually do with this?
Your hair look like poop, Bob! - Wanker.
D-waves systems are inherently statistical. Which means you need many replicas of an experiment to map out the ground state and reliably establish it is the ground state. Doesn't this mean that the more cubits you have the exponentially more replicas you need to run? thus anything short of exponential gains in speed is a step backward in perfromance as you add quibits? or am I wrong.
Some drink at the fountain of knowledge. Others just gargle.
Except to separate fools and their money. The one before was a bit faster than a simulation of itself on a slow, conventional computer. This one will still be massively slower than the best algorithm for calculating the same thing on a massively cheaper conventional computer. But many people are idiots, and some idiots have a lot of money, hence I do not doubt they will sell this SCAM-device as well.
Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
Ten hours on a bad day.
One cat dies for each bit that is settled in the solution.
Sheesh, evil *and* a jerk. -- Jade
...I can get a graphics card based on this, and a datajack?
[End Of Line]
Goddammit, I just found a qubit that needed annealing but I think its slipped down the back of the sofa.
But can it run Crysis?
But can it run crysis
And ECC. Probably not with this generation, but it's entirely possible three letter organizations are close to having a machine that can at least break smaller key sizes. Which in and of itself isn't especially worrying, except that the more sophisticated and rogue state-sponsored criminals won't be lagging very far behind.
There's no good reason why our web security infrastructure shouldn't immediately begin upgrades to support multiple, extensible and arbitrary methods of key exchange, including but not limited to stateful solutions using out of band preshared keys. Given it would only costs a few bucks to buy a gigabyte of key material stored on redundant robust flash memory, and each symmetric key wouldn't need to be more than a kilobit at the most, and in principle one would only need a couple keys to securely receive keys generated by trusted session key providing servers...
But what am I saying, this is an industry that had to scramble when somebody realized that the year 1999 was going to be followed by the year 2000. So it's probably far too optimistic to hope for an 'ounce of prevention' approach here.
I'm all for quantum computers. I think they will eventually become useful especially for biology yet their current and future capabilities like everything else are being way over hyped.
D-wave is not hundreds of billions of times faster neither does 1000 vs 2000 qubits marketing jargon mean anything substantive in and of itself.
To put this into perspective a real 2000 qubit quantum computer should be at least 10^300 times faster than a real 1000 qubit quantum computer. Not a measly billion or trillion or even a googol but a big honking 300 digit number times faster.
Quoting TFA:
"We should note that there are algorithms, such as techniques based on cluster finding, that can exploit the sparse qubit connectivity in the current generation of D-Wave processors and still solve our proof-of-principle problems faster than the current quantum hardware."
So are "Classical computers" the computers that Beethoven and Mozart used?
Before they go the way of Nortel/BNR, RIM, ... since they're Canadian eh! So am I so it's ok for me to say that eh!
You'll hear about it when real quantum computers reach commercial maturity, because a bunch of Slashdot articles will appear about how everyone is in a panic to rush and replace RSA with something else. :-)
"commercial maturity" being the key word here, because we should assume that significant portions of major classified intelligence budgets are being thrown at the problem by the US and China, maybe also by a few other players (India? Israel? The UK? Russia?). Like how it's widely believed that differential cryptanalysis was known to the NSA well before it became known to the world, only today encryption is much more prevalent and much more important to anyone doing signals analysis.
Real lawyers write in C++
I understand that what's being discussed is a device for simulated annealing, so perhaps the "programming" is part of the architecture. But will that be true for all future quantum devices? Or will they become more general purpose? If the latter, how does one go about programming a general-purpose quantum computer?
It's just marketing.
It's not quantum: There is no faster than light entanglement going on in there, and the qubits are not in superimposition state, they're in a state. It's not in superimposition state because it often doesn't get an optimum result. And the mechanism is electrical circuits and has no way to entangle.
What it is is an analogue computer using magnetic fields to solve Annealing problems, (optimizations that don't lend themselves to faster digital algorithms and instead have to be brute forced). You configure the circuit to represent the problem, cool it down till it superconducts and settles at a stable field representing *one* low energy solution, which has been found not to be the optimum solution in some tests.
It does not super-scale either. You cannot add 1000 qubits to that with zero overhead, so they'll let it settle for longer, turning down the gain slower, to compensate or perhaps it just gets shittier results. Who can tell without building a competing device to check the result against to see if it finds a better solution!
What competitors should do is build analogue computers in electronics to compete, because the magnetic field isn't essential for that, you could build one based on pure electronics. Such a device would be far cheaper and scale far better making it practical as a co-processor for annealing problems.
This is the best video explaining quantum computing I've ever found: https://www.youtube.com/watch?v=g_IaVepNDT4
"Some of the d-wave haters have moved onto the argument"....
Sure, its the haters.
@""Better" means searches a different solution space and therefore cannot solve all the same problems."
Yet if you have a problem that is better solved by a digital solver, why would you do it with a Dwave rather than just pull up the stock CMA-ES library, or Google deep dream code?
It's not a conspiracy of "haters" trying to keep DWave down, its mathematicians trying to explain to MBAs not to swallow hype like gullible idiots.
You start throwing words around like quantum and computer and annealing with massive levels of qubits - this is obviously where we are headed... right guys? RIGHT!?
Does it mean Windows would run at a decent speed on it?
Slashdot, fix the reply notifications... You won't get away with it...
... but the degree of entanglement.
It is in a superposition of actually working and not actually working.
"Last year, Google said that D-Wave's 1,000 qubit computer proved to be 100 million times faster than a classical computer with a single core"
But the quote you give next says not what you think it says. It is that much faster at annealing than simulating the same thing. The problem is that you will not simulate annealing to solve any real problem. They simulate the D-Wave system and oh wonder it was much slower than running a hardware implementation. That says nothing about the performance for any interesting problem.
That's like simulating every atom of my abacus on a supercomputer is maybe slower than using my abacus in real time. It does not mean a supercomputer computes slower than I with my abacus.
than conventional ones. It generates billions of random values in parallel, with incredible levels of entropy. I call it "cup of hot tea". I also invented a device capable of simulating three-body gravitational dynamics in real time, and I managed to reduce its power consumption to zero. It consists of two potatoes I throw off my roof. Sure, these are not general purpose computational devices yet, but I'm sure I can fix this in post.
It consists of two potatoes I throw off my roof.
Wow, you are a horrible god. You expect people to live on those potatoes?
Monsieur Garnier will be furious. He'd already bought the previous version for his laboratoire.
All you need now is an atomic vector plotter!
...is there no relevant XKCD? How will I ever be amused today?!
Especially for the Quantum Photo finish.
I present to you a formidable, all-new computer: An apple tree, with one ripe apple so loose it will fall off the tree soon.
You say you want proof that this new computer is superior to all classical computers? Here you go: I admit that my apple tree computer is optimized to solve a certain class of problems. It's optimized to simulate the behaviour of an apple falling down a tree - with insane precision, down to the level of considering even the smallest effects of quantum mechanics, gravity waves and neutrinos passing through the apple. Now go simulate that on your classical computer! And no cheating/simplification: You have to precisely factor in all those aspects of real physics as well!
You will find that my apple tree computer is finished solving the task while your Cray supercomputer is still busy calculating the bonding forces of the gluons and the quarks in the first few protons of your simulated apple! So my computer is a gazillion times faster than any classical computer at solving problem(s)!
This is how D-Wave advertises their hardware. They built some physical contraption resembling an analog computer and then claim that any classical computer is much slower, because the only task they are comparing it with is trying to simulate that very contraption they built.
Just marketing bullshit, which is so far from any useful advancement in technology.
All you need now is an atomic vector plotter!
I have a long plug and a small dangly bit.
Well I've wrestled with reality for thirty five years doctor, and I'm happy to say I finally won out over it.